<div dir="auto"><div><br><div class="gmail_extra"><br><div class="gmail_quote">On Sep 8, 2017 13:36, "Gandalf Corvotempesta" <<a href="mailto:gandalf.corvotempesta@gmail.com">gandalf.corvotempesta@gmail.com</a>> wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="quoted-text">2017-09-08 13:21 GMT+02:00 Pavel Szalbot <<a href="mailto:pavel.szalbot@gmail.com">pavel.szalbot@gmail.com</a>>:<br>
> Gandalf, isn't possible server hard-crash too much? I mean if reboot<br>
> reliably kills the VM, there is no doubt network crash or poweroff<br>
> will as well.<br>
<br>
</div>IIUP, the only way to keep I/O running is to gracefully exiting glusterfsd.<br></blockquote></div></div></div><div dir="auto"><br></div><div dir="auto">No, even killall resp. SIGTERMing glusterfsd ends up with I/O errors on the client. </div><div dir="auto"><br></div><div dir="auto">I did not test SIGKILL because I suppose if graceful exit is bad, SIGKILL will be as well. This assumption might be wrong. So I will test it. It would be interesting to see client to work in case of crash (SIGKILL) and not in case of graceful exit of glusterfsd.</div><div dir="auto"><br></div><div dir="auto"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
killall should send signal 15 (SIGTERM) to the process, maybe a bug in<br>
signal management<br>
on gluster side? Because kernel is already telling glusterfsd to exit,<br>
though signal 15 but glusterfsd<br>
seems to handle this in a bad way.<br>
<br>
a server hard-crash doesn't send any signal. I think this could be<br>
also similiar to SIGKILL (9)<br>
that can't be catched/ignored software side.</blockquote></div></div></div><div dir="auto"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
In other words: is this a bug in gluster's signal management (if<br>
SIGKILL is working and SIGTERM no, i'll almost sure this is a bug in<br>
signal management),<br>
a engineering bug (relying only on a graceful exit [but even SIGTERM<br>
should be threthed as graceful exit] to preserve I/O on clients) or<br>
something else ?</blockquote></div></div></div><div dir="auto"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
</blockquote></div><br></div></div></div>